Gold plaque of a barsom-bearer (Oxus Treasure)

The plaque is a rectangle of gold about 15 cm high and 7.5 cm wide, and barely a millimetre thick, the figure raised in low relief by embossing from behind (repoussé) and sharpened from the front with a chasing tool.^[1] Scientific analysis gives the metal as gold 93 percent, with a little silver and copper, the ordinary high-gold alloy of the treasure's finer pieces.^[1] It was catalogued by O. M. Dalton as no. 48 of the treasure, illustrated on his plate XIV, in the standard catalogue that has framed every later discussion.^[2]

The figure faces right, his hair showing above and below the hood, his tunic marked with vertical bands of applied or embroidered ornament and further bands at the wrist and shoulder. This is the dress, the British Museum notes, of the tribute-peoples shown coming from the north or north-east on the western façade of the Palace of Darius at Persepolis, and it is matched by an actual costume, cap, tunic, trousers and all, recovered from the frozen Issyk kurgan far out on the Saka steppe, so that the plaque's "Median" costume is really the shared riding-dress of the northern Iranians.^[1] The akinakes at the belt is the same short sword the reliefs hang on Persian and Median nobles alike. Everything on the figure, then, is a known Achaemenid type; only the bundle in his hand carries the object into the world of the cult.

The plaque is the best-finished member of a series, not a unique masterpiece. The treasure held about fifty of these thin gold sheets, catalogued by Dalton and re-studied by John Curtis and Ann Searight, who counted a set of some fifty-two.^[3] They range widely in size and skill, from little strips a few centimetres high, their figures crudely scratched in, to this larger, carefully modelled sheet; the subjects run from men in Median or Persian dress carrying the barsom or a flower, through women in long robes, to a horse and the foreparts of a camel.^[4] Cut from cheap thin sheet, repeated in numbers, and worked with every grade of care, they are plainly not portraits but dedications, standardised tokens left at a holy place. The Forgotten Empire catalogue states the reading plainly:

These plaques are clearly votive, and they and the other objects in the treasure have the appearance of material that was dedicated to a temple perhaps over a period of several centuries.

Curtis and Tallis, Forgotten Empire^[4]

The same devotion appears in the round in a handful of small votive statuettes from the treasure: a silver figure of a bearded man in Persian robe and crown holding a short barsom to his breast, and a gold figure in Median dress with the tiara stiffened to stand erect, both, Boyce notes, presumably kings by the royal manner of the tiara, and both clasping the rods.^[5] Of the plaques, Boyce singled out the British Museum sheet as the finest:

The finest of these shows a man in tunic and trousers but without the kandys ... in his right he holds an impressively long baresman, and at his side he wears the characteristic Iranian short sword.

Boyce, A History of Zoroastrianism II^[5]

The kandys Boyce marks as missing is the Median riding-coat. Cheap, repeated dedications are the mark of a popular and habitual rite, not a rare royal one: ordinary worshippers, or people acting for them, left these tokens at a shrine as a matter of course, the way later ages left wax limbs or painted panels. The barsom rite the plaque shows was thus not a court ceremony glimpsed once but a common act of devotion, performed often enough to sustain a workshop turning out gold figures of the worshipper at prayer. The one plaque that reproduces best is a window onto a practice attested by the fifty around it.

The bundle the figure holds is the heart of the matter. The barsom (Middle Persian for the Avestan barəsman, from a root meaning "to grow high") is a set of sacred twigs that forms, in the words of the Encyclopaedia Iranica, "an important part of the Zoroastrian liturgical apparatus".^[6] In the old rite the twigs were of a living plant, the pomegranate or another, cut with prescribed prayer, laid out, and tied into a bundle, the number fixed by the ceremony to be performed. The celebration of the Yasna, the central act of worship, required twenty-three twigs, of which twenty-one made the bundle; the longer Vendidad service took thirty-five.^[6] The priest held the barsom in his hand while he chanted the liturgy, and the twigs rested on a pair of crescent-topped metal stands, the barsomdān. In later, temple-based practice the living twigs gave way to thin rods of brass or silver, each about nine inches long, still called by the old name; it is this metal form that the Sasanian and modern Zoroastrian priest uses, and that appears in the hands of later divine and royal figures in Iranian art.^[6]

What the bundle meant is spelled out in the rite itself. The Vendidad has Zarathustra ask Ahura Mazdā how the god may fitly be praised, and the answer sends the worshipper to a growing tree:

Go near a tree grown out of the earth and repeat thus: 'Homage unto thee, O beautiful, flourishing, strong and Mazdā-created tree.'

Vendidad 19.17–18, trans. Darmesteter^[7]

The twigs, cut with that homage, stand for the whole of the vegetable creation, and the holding of them is a thanksgiving to the creator for the produce of the earth on which human and animal life depends. The barsom rite is an act of gratitude to the maker of the good world, the same theology the kings carved on their tombs, performed with a handful of rods. Its ritual home is the fire-service and the wider Magian cult treated on their own entries; on the plaque it is reduced to its one legible gesture, the bundle raised before the body.

The plaque puts a body and a gesture to the barsom, and two Greek witnesses give the scene an independent check. Herodotus, describing Persian sacrifice, insists on the priest at its centre:

A Magian man stands by them and chants over them a theogony (for of this nature they say that their incantation is), seeing that without a Magian it is not lawful for them to make sacrifices.

Herodotus, Histories 1.132, trans. Macaulay^[8]

Strabo, drawing on Hellenistic sources, is more exact still, and describes almost precisely the figure the plaque shows, down to the bundle of rods and the felt cap drawn over the face:

In the midst of these there is an altar, on which there is a large quantity of ashes and where the Magi keep the fire ever burning. And there, entering daily, they make incantations for about an hour, holding before the fire their bundles of rods and wearing round their heads high turbans of felt, which reach down over their cheeks far enough to cover their lips.

Strabo, Geography 15.3.15, trans. Jones^[9]

The "bundles of rods" are the barsom. Albert de Jong, whose commentary is the standard modern reading of the Greek and Latin evidence for Persian religion, takes Strabo's slender wands and small sticks for the object without hesitation: "This refers to the baresman (Phl. barsom), the bundle of sacred twigs."^[10] The felt "turbans" drawn over cheeks and lips are the hood on the plaque, the padām or mouth-cover that keeps the priest's breath from defiling the holy things; the same usage survives among Zoroastrian priests today, who cover the mouth with a cloth of that name.^[4] For the same reason, Strabo adds nearby, the fire itself must never be breathed upon:

To fire they offer sacrifice by adding dry wood without the bark and by placing fat on top of it; and then they pour oil upon it and light it below, not blowing with their breath, but fanning it; and those who blow the fire with their breath or put anything dead or filthy upon it are put to death.

Strabo, Geography 15.3.14, trans. Jones^[11]

So the plaque, a Greek geographer four centuries later, and the liturgical books agree on the same scene: a figure in Iranian dress, mouth covered, holding a bundle of rods before him in worship. That agreement across independent kinds of evidence, an object from Bactria, a text from Cappadocia, a liturgy from Iran, is what gives the little plaque its weight.

Whether the man on the plaque is a priest (a magus), a lay worshipper in the priestly posture, or the donor who dedicated the plaque, is not settled, and the object cannot decide it. Three readings are in play, each resting on a real feature of the image.

The first takes him for a priest. The barsom and the mouth-covering hood are the officiant's apparatus in the later liturgy, and the British Museum's exhibition scholarship reads the barsom-carrying plaques as showing priests in Median costume with padām and barsom, matching the fifth-century relief from Daskyleion where two such figures, mouths covered, hold bundles of rods before a shrine with the heads of sacrificed animals.^[4] Shahrokh Razmjou, arguing from Persepolis, adds the priestly markers Greek and later sources supply: the clean-shaven face, the drawn-up chin-and-mouth cover, and the white dress that Boyce inferred for priests from the faded colours, so that the barsom-bearer is the officiant of the fire-cult.^[4] On the Daskyleion relief one of the two rod-bearers is clearly beardless, and a shaven face was a priestly sign across the Near East, for Egyptian, Assyrian and Elamite clergy alike, which strengthens the priestly reading of that scene; but the little Oxus figures are too summary to carry the same inference, and several of them are bearded.^[4]

The second reading resists this. The figure wears the akinakes, and a sword is a warrior's, not a priest's; whether priests ever carried one, Boyce concluded, there is not enough parallel evidence to say, so the man may be a nobleman shown in a devotional attitude rather than a magus.^[5] The votive statuettes sharpen the doubt, for the tiara worn erect on them was a royal privilege, which would make those figures kings at prayer. In the earliest usage the barsom was not the priest's alone: the Encyclopaedia Iranica notes that it seems to have been required even for the ordinary worshipper's simple grace before meals, held in the hand during the recital, so a layman at prayer would also have carried it.^[6] The Ferdowsi story of the last Sasanian king, Yazdegerd, calling for a barsom to say grace while in hiding (his asking for it betraying him to his killers) shows the object in a king's hands.^[6]

The British Museum's own catalogue keeps the question open in a way that points to the third reading, allowing that "the figures carrying barsoms might be priests, or the barsoms might be indications of piety".^[4] On this view the distinction between priest and layman is beside the point: a votive shows an act of devotion, and the barsom is its badge, marking the figure, whoever he is, as one at worship. The plaque, showing a single generic figure in the posture of prayer, is compatible with all three readings, and the safest description is the one the object itself supports, an Iranian at the barsom rite. The plaque is no portrait of a particular man: any individualising of the face sits on an otherwise generic votive type. It is a dedicatory image of worship as such, and its very indifference to who is shown is what makes it good evidence for the rite rather than for a person.

The plaque's chief difficulty is the hoard it belongs to; the figure itself is not in doubt. The Oxus Treasure (some 180 surviving gold and silver objects, with a body of associated coins) is the largest single group of Achaemenid precious metalwork known, and it includes some of the most famous images of the empire: a pair of griffin-headed gold armlets, a small gold model chariot drawn by four horses with two riders in Median dress, gold and silver statuettes, vessels, sword-fittings, seals, and the plaques. None of it was excavated. It surfaced on the antiquities market between about 1876 and 1880, said to have been found on the north bank of the Oxus (the Amu Darya) and allegedly scattered in the sands of the river, and it passed through the hands of local merchants toward the bazaars of British India before it reached England.^[4]

A romance attached itself to the find, and it is a caution in itself. As the story is usually told, merchants carrying the treasure were set upon by bandits and a British political officer recovered the goods by night from the robbers' cave; what is documented under the tale is soberer and thinner. Boyce records only that "after stirring incidents most of them reached the markets of Rawalpindi, where their number was apparently augmented by some ingenious fakes, and where various other objects seem to have become associated with them".^[5] One fixed point survives: Captain F. C. Burton bought the great gold armlet in 1884 (it is now in the Victoria and Albert Museum). The greater part came to the British Museum in 1897 with the bequest of Sir Augustus Wollaston Franks, who had acquired it with Sir Alexander Cunningham.^[4] The colourful details cannot be checked, and the essential facts they dress up are damaging: the treasure was assembled on the market, its numbers swelled by fakes and by objects that drifted into the group, its origin only reported and never observed.

Two questions have been argued ever since, and they must be distinguished. The first is authenticity, and here the verdict is reassuring. John Curtis's assessment for the museum is that claims that pieces are faked "are not well founded", and that scientific examination, art-historical analysis and the emergence of new parallels "all point to its authenticity".^[4] The plaques and the major objects are genuine Achaemenid work, whatever the market did to their edges. The second question, provenance and coherence, is the one that will not close, and it governs how far the plaque can be pressed as evidence.

If the plaques are votive, they were dedicated somewhere, and where has become the treasure's central puzzle. The British Museum gives the findspot of this plaque as Takht-i Kuwad (Takht-e Qobād), a fortress and ferry-station on the river, about a quarter of which has been washed away; Curtis's judgement is measured, that "the question of provenance is more problematic, but Takht-i Kuwad remains a possibility".^[4] Five kilometres upstream lies a more tempting candidate. At Takht-i Sangin, Russian excavations between 1976 and 1991 uncovered a temple on a citadel whose storerooms held more than five thousand objects dated between the sixth century BCE and the third century CE, the so-called Temple of the Oxus.^[4] The discovery has drawn the treasure toward it, and the Encyclopaedia Iranica records the pull: the hoard is "thought by some to have originated as votives in the Temple of the Oxus, but may have come from somewhere else in the vicinity, such as the earlier fortress site downstream at Takht-e Qobād".^[12]

The temple hypothesis is attractive and unproven, and Curtis lists the obstacles. The Takht-i Sangin temple was built at the very end of the Achaemenid period or a little later, whereas the plaques are earlier; and the three gold votive plaques actually found at Takht-i Sangin are of a different character from those in the treasure, so the sanctuary that certainly received votives is not obviously the one that received these.^[4] The coins reported with the hoard, running down to about 200 BCE, would date any single burial to the Hellenistic period, long after the plaques were made, which fits a shrine's treasury accumulated and finally hidden over centuries better than a single act of deposit.^[4] Whether the treasure is one deposit or a dealer's assemblage of finds from more than one place is therefore still open. Ellen Rehm's survey notes that Muscarella questioned the treasure's reported Oxus origin and that whether all the objects belonged together remains under discussion.^[13] Nothing here rests on a controlled excavation, and the dating of the plaque is broad by necessity; the British Museum assigns this one to the fifth or fourth century BCE, and no claim resting on the hoard can rise above the probable.^[1]

The deposit question bears on reading the plaque as much as on placing it. If the treasure was a shrine's accumulated treasury, buried in the Hellenistic period after generations of gifts, then the plaques were what they look like, votives left by worshippers, and the object's own form is confirmed by its context. If instead it was a hoard swept together and hidden, or a dealer's group assembled from several finds, then the votive reading rests on the plaque alone, on its thinness, its repetition across a series, and the devotional posture of its figure, without the temple to vouch for it. The evidence happens to point the same way from both ends: the objects look dedicated whether or not their shrine is ever found. But the warranted claim is narrow. The plaque is a genuine Achaemenid image of the barsom rite; it is probably a votive; and it comes from Bactria in the later empire. Beyond that the archaeology gives no warrant, and the caution is no formality; it marks the difference between what the object shows and what the hoard cannot.

Whatever its exact deposit, the treasure is a Bactrian object, from the empire's north-eastern march, the satrapy on the Oxus, and this plaque may be Bactrian in its making as well as its findspot. The British Museum records that its sheet is thicker, and its workmanship of a "considerably different standard", than the other plaques of the treasure, and reports the view of Lolita Nehru that it is possibly evidence of local Bactrian production: the face is somewhat individualised, with a faint trace of portraiture, unlike the highly stylised, formulaic figures on the Persepolis façades.^[1] If so, the object is doubly valuable: a provincial hand, out at the edge of the Iranian world, showing the same barsom rite the imperial centre and the Greek witnesses describe.

The comparison with the heartland is instructive, and it runs the plaque's way. The Persepolis archives preserve their own fire-cult scenes, cut on the seals of the administrators: a figure stands at a stepped fire-structure, extending a vessel or, it seems, a bundle of rods over the flame. But there the identification is tentative. Mark Garrison, cataloguing the seal imagery, notes that the objects held above the fires may be bundles "potentially connected to the often-discussed barsom", yet are so poorly preserved in the tiny impressions that "one is reluctant to infer specific identifications for them".^[14] The seals show the rite equivocally, at the scale of a fingernail; the Bactrian plaque shows it plainly, at the scale of a hand. The eastern object is clearer evidence for the barsom than anything from the centre, which is a large part of why it is so often reproduced.

That is the plaque's real historical yield. Achaemenid religion is known mostly from Persis and from the Greek writers who watched the court; here is direct evidence that the worship attested for the Magi, for Herodotus, and for the Avesta was practised across the whole span of the empire, from the heartland to the far Oxus, in the same forms and with the same furniture. The dahyāva (the lands and peoples) of the royal lists were held together by more than tribute; a common cult reached to Bactria, and a gold figure with a bundle of rods in his hand is one of the few things that shows it there.

The barsom outlived the empire, and changed form as it did. In the Achaemenid rite it was a bundle of living twigs, cut with prayer from a growing plant; but the demands of temple worship and the enclosed fire-cult of later Zoroastrianism turned it, in time, into a set of thin metal rods, brass or silver wires, each about nine inches long, laid up in the requisite number and handled with the old prescribed formulae.^[6] The rite survives in living Zoroastrian practice, where the priest still consecrates the barsom and touches the ends of the bundle with the water-wire in the appointed order.^[6]

The metal barsom also entered the picture of divinity itself. Where the Achaemenid god had no image at all, the later Iranian world came to show its great figures bearing the rods: in Sasanian investiture reliefs Ahura Mazdā, by then Ohrmazd, can appear as a crowned figure holding the barsom, the priestly emblem raised to the divine hand, the same object the anonymous Bactrian on the gold plaque holds. The handful of rods is one of the very few ritual objects that can be traced, in unbroken use and with its name intact, from the Avestan liturgy through the Achaemenid rite to the fire-temples of today. The plaque stands near the beginning of that long continuity, a votive from the edge of Darius's empire at the head of a line that runs to the iconography of the Persian kings a thousand years later.